We therefore decided to re-read,post-hoc but in a blinded fashion,the digital
radiographic images also according to a semi-automated technique in agreement with current guidelines to increase precision, and to limit the analysis to the medial tibiofemoralcompartment, in order to better substantiate the previous results.Actually,there
was no statistical interaction between treatments with either the
semi-automated or the visual(manual) measurement of mJSN in the medial compartment.Indeed,the results were very comparable for
the two methods,with asignificant decrease of 58–59% in the risk of
developing knee OA in the women who had received glucosamine
sulfate with or without DEP.